ClawHub
Back to skill

Security audit

Crypto First Steps Coach

Security checks for vulnerabilities and agentic risk

Overview

This is a prompt-only beginner cryptocurrency education skill, and the extra code only confirms that it does not read local files.

This skill is low risk from the inspected artifacts. It is still about cryptocurrency, so users should treat it as educational guidance only and not as financial advice or a tool for choosing investments.

A.I.G

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs

SkillSpector

Vulnerability Patterns
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (1)

Tp4

High
Category
MCP Tool Poisoning
Confidence
95% confidence
Finding

The declared purpose is a calm educational crypto onboarding guide. The actual code chunk does not implement or demonstrate that behavior; instead, it tests whether a handler safely resists local file access when given a path-like skill name. That is a materially different primary purpose and indicates security/file-access behavior not reflected in the description. Even though the test expects no local file read, the capability area under examination is unrelated to the declared crypto-guide functionality.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.